Why did some practices not implement new antibiotic prescribing guidelines on urinary tract infection? A cohort study and survey in NHS England primary care
Croker R., Walker A., Goldacre B.
Abstract Objectives To describe prescribing trends and geographic variation for trimethoprim and nitrofurantoin; to describe variation in implementing guideline change; and to compare actions taken to reduce trimethoprim use in high- and low-using Clinical Commissioning Groups (CCGs). Design A retrospective cohort study and interrupted time series analysis in English NHS primary care prescribing data; complemented by information obtained through Freedom of Information Act requests to CCGs. The main outcome measures were: variation in practice and CCG prescribing ratios geographically and over time, including an interrupted time-series; and responses to Freedom of Information requests. Results The amount of trimethoprim prescribed, as a proportion of nitrofurantoin and trimethoprim combined, remained stable and high until 2014, then fell gradually to below 50% in 2017; this reduction was more rapid following the introduction of the Quality Premium. There was substantial variation in the speed of change between CCGs. As of April 2017, for the 10 worst CCGs (with the highest trimethoprim ratios): 9 still had trimethoprim as first line treatment for uncomplicated UTI (one CCG had no formulary); none had active work plans to facilitate change in prescribing behaviour away from trimethoprim; and none had implemented an incentive scheme for change in prescribing behaviour. For the 10 best CCGs: 2 still had trimethoprim as first line treatment (all CCGs had a formulary); 5 (out of 7 who answered this question) had active work plans to facilitate change in prescribing behaviour away from trimethoprim; and 5 (out of 10 responding) had implemented an incentive scheme for change in prescribing behaviour. 9 of the best 10 CCGs reported at least one of: formulary change, work plan, or incentive scheme. None of the worst 10 CCGs did so. Conclusions Many CCGs failed to implement an important change in antibiotic prescribing guidance; and report strong evidence suggesting that CCGs with minimal prescribing change did little to implement the new guidance. We strongly recommend a national programme of training and accreditation for medicines optimisation pharmacists; and remedial action for CCGs that fail to implement guidance; with all materials and data shared publicly for both such activities.